1. Field of the Invention
This invention relates to a digital video signal recording/reproducing apparatus using bit rate reduction coding and having a function of outputting reproduced data in a form of a bit rate reduction coded state.
2. Description of the Prior Art
Under a progressive recent trend toward making high quality, digital home-use video equipment, video signal recording and reproducing equipment using digital recording technology have been developed. However, if an input video signal is a 4:2:2 component signal, for example, the transmission rate becomes as high as 216 Mbps, and if it is directly recorded, a large recording capacity for allowing a long-time recording cannot be realized. As a result, in order to obtain an adequately-long recording time for home-use, it is required to efficiently reduce the quantity of image data to a degree capable of obtaining allowable picture quality by bit rate reduction coding technology. The structure and operation of a conventional digital video signal recording and reproducing apparatus using bit rate reduction coding technology is described below.
First, when recording, an input video signal is converted into a digital signal by an A/D converter and compressed into a predetermined quantity of data by a bit rate reduction encoder. As a method of bit rate reduction encoding the video signal, a method of using an orthogonal transformation is generally employed from the viewpoint of compression efficiency. The orthogonal transformation divides plural pixels into blocks and converts the pixels into frequency regions on a block by block basis. By encoding the transformed coefficients by variable length encoding, the data can be efficiently compressed. In this case, if the unit for the variable length encoding is increased, the efficiency of encoding can be improved. 0n the other hand, if an error is generated, the range of error propagation becomes large. As a result, in the digital video signal recording and reproducing apparatus having a higher probability of error generation than other transmission lines, an encoding method in which the bit rate reduction encoding is performed in a unit of plural blocks (hereinafter called a compression block) has been introduced.
Next, in an error correction encoder, an error correction parity coded is added, and then a sync signal and identification data (ID) are added to form a recording block (hereinafter called a sync block). Then, the sync block is modulated for recording and is recorded on a recording tape. The sync block is essentially composed of a sync signal for detecting the sync block from the reproduced signal, identification data (ID) containing a track number for correctly writing the reproduced sync block into a memory and a sync block number, bit rate reduction encoded data of the video signal, and the error correction parity code. The sync blocks thus composed are recorded on a track on the tape.
Next, when reproducing, a reproduced signal from a reproducing head is subjected to detection and demodulation of the sync block, and subjected to error correction in an error correction decoder according to the error correction parity code added during recording. Error correction is effected up to a certain number of errors by this process. However, if a dropout occurred, for example, due to tape damage such dropout error that cannot be corrected by the error correction decoder. If an uncorrectable error occurred as such, error concealment is performed so that the influence of the error can be as inconspicuous as possible visually. The influence of the error propagates to the range within which a variable length encoding is completed. As a result, in an error concealment circuit, error concealment is carried out by substituting an entire sync block forming a compression block containing erroneous data with a previous corresponding sync block. The concealed data is decoded into the original data in a bit rate reduction decoder and converted through a D/A converter into the analog signal. A digital recording VTR using the bit rate reduction coding technology as shown above is disclosed, for example, in "An Experimental Study on a Home-Use Digital VTR" by C. Yamamitsu et al., IEEE Transaction on Consumer Electronics, Vol. 35, No. 3, August 1989, pp. 450-457.
In the digital recording, a data sequence of 0 and 1 is recorded. If 0 and 1 are detected erroneously due to the distortion of the reproduced waveform, it becomes an error of the data. In this case, if the error rate is a predetermined value or less, the error can be corrected by error correction techniques. Therefore, by setting the correctability adequately with respect to the error rate of the reproduced data, perfectly correct data can be reproduced. As a result, contrary to the analog recording in which repetition of dubbing results in an increase and accumulation of distortion, the digital recording is largely advantageous in that deterioration due to dubbing can be largely reduced. Here, "dubbing" means that a signal reproduced from one reproducing apparatus is sent to and recorded by another recording apparatus. Particularly, in digital dubbing using no analog processing, deterioration of the signal can be further largely reduced.
Here, when the orthogonal transform encoding is used as a method of bit rate reduction encoding of a video signal, the orthogonal transformation and inverse orthogonal transformation are actually processed in a finite word length, so that the repetition of the orthogonal and inverse orthogonal transformations results in an accumulation of operational errors, leading to an increase in distortion. Besides, when that filtering processes such as decimation, interpolation and the like are additionally carried out, the repetition of such filtering processes results in an increase in distortion as well. As a result, though the digital recording is advantageous in that deterioration due to dubbing is small compared with the analog recording, the distortion will be increased whenever the digital dubbing is carried out, so that picture quality after the dubbing has been carried out several times may be deteriorated to such a degree that it can be visually sensedw compared with the original picture quality. Accordingly, in order to prevent deterioration due to dubbing from being accumulated, a method of dubbing signals in the bit rate reduction encoded state is disclosed, for example, in U.S. Pat. No. 5,193,010.
However, the method disclosed therein is such that concealed data are outputted for dubbing in any case. As a result, it cannot respond to any case where error concealment is not carried out in the bit rate reduction encoded state. Also, even when the concealment is to be carried out, it must be done in the form that the decoding is correctly carried out in the bit rate reduction decoder.